This is the first instalment of a series about dfdao’s journey in Dark Forest Round v0.6.5. dfdao is a collective of gamers, builders, and artists who love Dark Forest, the premier blockchain game. I’m 0xhank, dfdao’s chief historian and contributor. If you want to help support more content like this from dfdao, please donate to our Gitcoin grant round.
Our tale begins where all good web3 stories start: on a Discord voice chat. My friend Tony and I were planning a Dark Forest hackathon led by dfdao. He and I had been friends through college, where we spent our time coding and playing ultimate frisbee. After graduating, we stumbled across web3 thanks to his uncle and dfdao co-founder, Will Robinson. Tony is an avid reader, passionate political thinker, and obsessive coder. And now he is cha0sg0d, the co-founder of dfdao and its primary coordinator. As for me, I’m 0xhank, and I contribute to dfdao and write about web3.
Dark Forest is unique because to get a high score, you need more than just gaming skills — you must strategize, perform diplomacy, and, most importantly, write code. This is because the game allows players to write JavaScript-based plugins to manipulate its backend, which is built entirely on smart contracts and deployed to the Gnosis blockchain. dfdao’s guild was founded by Will in the summer of 2021 so he could win despite having limited technical skills. Read about his experience here. dfdao has since evolved into a thriving community and a great place to build.
Surprise! A New, Extra-Long Dark Forest Round
Just as Tony and I were about to post the introductory Twitter thread to our hackathon, the Dark Forest core team announced the first official round in three months. This round would be extra long, lasting two straight weeks without pause. As soon as we saw the post, we knew our hackathon was going to be postponed. We needed to spend the next few weeks creating and executing a plan to topple the three-time champion OrdenGG.
The new Dark Forest round was called “Junk Wars.” There were three new mechanics: space junk, capture zones, and spaceships. For those who don’t know what Dark Forest is, go read about it here. Here is how each new mechanic works:
Space Junk: Every time you capture a new planet, you accrue junk. When your junk hits 2000, you can’t capture any more planets. You can abandon a planet to lose its space junk.
Capture Zones: Throughout the map, there are randomly generated capture zones that change every 20 minutes. If you invade a planet within a capture zone and hold it, you can capture it and win points.
Spaceships: Five special artifacts that, when docked on a planet, affect the planet in various ways. The two most important ships are the Mothership and Whale, which respectively double energy regeneration and double silver regeneration.
If you want more details about these new mechanics, check out the Dark Forest blog post explaining the new rules.
Putting together the Team
OrdenGG is a tight-knit guild of programmers and skilled players. Their reputation precedes them; they won the last three rounds of Dark Forest v0.6, often by a wide margin. Before this round, they jokingly goaded other players with a tweet of a huge (and expertly rolled) blunt. While the trash talk was in good fun, we knew their reign needed to end.
After months of player initiatives, like the Death of the Universe and Learn to Play, dfdao has built a thriving community of Dark Forest players and builders. We had access to talented people — we just had to recruit them.
Our Spanish friend Velorum was a no-brainer — he leads community management for dfdao, is a skilled pilot and, as an obsessive young man like ourselves, was prepared to sacrifice the next two weeks of his life to win. He and Tony would lead the guild for the next two weeks.
Velorum and Tony reached out to skilled web3 developers in the dfdao community who love the technical side of Dark Forest. David, aka Nifty Maestro, is a Vietnam-based Brit with years of programming experience. Doug Binder runs a small consulting firm focused on FinTech/Ethereum and has open-sourced a handful of Dark Forest plugins. And Francesco Ceccon is a programming mastermind in the StarkNet universe who offered to help out.
In addition to technical players, we found a set of passionate all-rounders in dfdao to help however they could. MJ is a player since v0.3 known for his math-focused plugins, Bulmenisaurus is a mysterious community member and Modukon is an Austrian who built plugins for the community round. Others offered mining rigs for exploration: Waffle, a Galcon legend who joined during the Learn to Play round, and Wombocombo, a GPU god.
As we had no money, all we could offer our contributors was a share in the governance of our potential reward planet. More importantly, we proposed a chance for glory. Our hodge-podge team had programming skills, computing power, strategic thinking, and, most importantly, motivation. We had all the tools to take the top prize — now we just had to execute.
The Strategizing Process
Access to a staging round allowed us to experiment with new game mechanics. The core Dark Forest team uses the staging round to let community members, including Orden, and a Chinese guild called MarrowDAO, find and report bugs prior to the round’s release. After clearing it with the core devs, Bulmenisaurus and Velorum used the staging world scoring rules to create a round strategy.
Once the testing phase was out of the way, Velorum spearheaded a massive Excel document based on the world constants of the staging round. He calculated the number of planets a player could capture before they ran out of space junk, the expected number of planets in the game world, the amount of silver an asteroid field produced, the speed and distance of different planets, and the effect spaceships would have.
Using our spreadsheets and staging round experience, we formulated a strategy to win the round.
Our Initial Strategy
To beat OrdenGG, we knew we had to utilize our advantage of pure numbers, with ten guild members and additional support in the broader dfdao community. This would let us delegate skilled tasks, deploy hordes of spaceships to strategic locations, and ask for outside help if things got desperate.
Our research unveiled a quirk we could use to our advantage: the Whale spaceship allows an asteroid to double its silver regeneration. If we stacked 16 Whales on an asteroid field, we could boost silver by production by a factor of 2¹⁶. Velorum calculated that this whale farm, which we named “Moby Dick,” would produce nine million silver per hour; more than enough to win the game for us. We went into the round optimistic that OrdenGG would finally lose.
Little did we know that using the staging round to strategize would massively backfire. The value of silver in the staging round was radically changed for the actual game, so the Moby Dick was far weaker than we initially calculated. We would end up stuck at hundreds of millions of points behind Orden, with no clear path to victory… but that, my friends, is a story for next time.
